Why are there so many people in aurodium with 5 GLs

Replies

  • Dawnsinger wrote: »
    Drathuk916 wrote: »
    Drathuk916 wrote: »
    The squish is the answer. Instead of audorium being players ranked in the 60-80 percentile, as cg has said their goal is, it contains players in the 68-84 percentile.

    So that’s my original quote. To expand on it, cg stated in the beginning that their goal was to have 5 leagues evenly balanced in population. The individual division were targeted to be split at 10/25/30/25/10. The overall population has been relatively stable at 300-310k for a while which means if cg was meeting its stated goal each league should look like 6000/15000/18000/15000/6000.

    Now no league comes close to looking like that and it mainly has to do with the squish artificially forcing players towards division three. It’s true kyber has somewhat stabilized and you do have to play your way out of kyber vs being squished out. However, without spending too much time studying the math, it feels like the squish has made movement between leagues far more difficult than it should be. My gut instinct seems to be supported by the shrinking to almost disappearing entirely of the populations in division 1 and 5 in any league not just kyber.

    My apologies. I assumed you were the guy I replied to, I didn’t realize you were using my post to talk about a totally new (but admittedly interesting) topic.

    There definitely don’t appear to be using the squish to hone in on their initial targets for the size of each division. It is being used almost exclusively to shift players up from Carbonite and into the other leagues until you get to Aurodium. Without it the size of Bronzium would be absolutely mammoth. If anything they’re not being anyway near aggressive enough with the squish as evidence by the fact that Carbonite has grown so big it’s shifted all the leagues away from their stated targets. There just are too many people entering Carbonite every month and they aren’t shifting them out fast enough to keep up.

    If you do the math you’ll see it’s actually made it much easier to move up the leagues with the exception of Aurodium the past 6 months, but if you’re in Aurodium I can see why your gut would say it’s harder.

    I’m in k2. My personal experience is I don’t care cause I’m getting more crystals than I need to progress at the rate I desire. That said I feel the squish is too punishing in k2. You need to average around 5.2 wins per season to maintain your rating. I’d lower that to somewhere between 4.7-5 wins. I’m 19-11 over my past 30 matches and gained around 130-150 rating points that’s too low.

    Finally, I’m not entirely clear why you feel it’s easy to climb so I must be missing something. It looks like the three middle leagues have shrunk by at least 12 percent while carbonite has grown by over 33%.
  • Drathuk916 wrote: »
    Dawnsinger wrote: »
    Drathuk916 wrote: »
    Drathuk916 wrote: »
    The squish is the answer. Instead of audorium being players ranked in the 60-80 percentile, as cg has said their goal is, it contains players in the 68-84 percentile.

    So that’s my original quote. To expand on it, cg stated in the beginning that their goal was to have 5 leagues evenly balanced in population. The individual division were targeted to be split at 10/25/30/25/10. The overall population has been relatively stable at 300-310k for a while which means if cg was meeting its stated goal each league should look like 6000/15000/18000/15000/6000.

    Now no league comes close to looking like that and it mainly has to do with the squish artificially forcing players towards division three. It’s true kyber has somewhat stabilized and you do have to play your way out of kyber vs being squished out. However, without spending too much time studying the math, it feels like the squish has made movement between leagues far more difficult than it should be. My gut instinct seems to be supported by the shrinking to almost disappearing entirely of the populations in division 1 and 5 in any league not just kyber.

    My apologies. I assumed you were the guy I replied to, I didn’t realize you were using my post to talk about a totally new (but admittedly interesting) topic.

    There definitely don’t appear to be using the squish to hone in on their initial targets for the size of each division. It is being used almost exclusively to shift players up from Carbonite and into the other leagues until you get to Aurodium. Without it the size of Bronzium would be absolutely mammoth. If anything they’re not being anyway near aggressive enough with the squish as evidence by the fact that Carbonite has grown so big it’s shifted all the leagues away from their stated targets. There just are too many people entering Carbonite every month and they aren’t shifting them out fast enough to keep up.

    If you do the math you’ll see it’s actually made it much easier to move up the leagues with the exception of Aurodium the past 6 months, but if you’re in Aurodium I can see why your gut would say it’s harder.

    I’m in k2. My personal experience is I don’t care cause I’m getting more crystals than I need to progress at the rate I desire. That said I feel the squish is too punishing in k2. You need to average around 5.2 wins per season to maintain your rating. I’d lower that to somewhere between 4.7-5 wins. I’m 19-11 over my past 30 matches and gained around 130-150 rating points that’s too low.

    Finally, I’m not entirely clear why you feel it’s easy to climb so I must be missing something. It looks like the three middle leagues have shrunk by at least 12 percent while carbonite has grown by over 33%.

    If you want to see the details you should check out the excellent work Taliana has done to show the details. Just search her name on the forums. But the short version is that Carbonite gets a massive boost every seasons from the squish (100s of extra SR) Bronzium gets a large boost and Chromium gets a small one. The reason they have to do this, and why Carbonite is still growing compare to the rest is that people are constant leaving and new players are joining. But they leave from all over and only join in Carbonite, so the lowest league is naturally getting bigger while everywhere else is shrinking.

    The squish definitely makes it tough to stay in high K2 or K1 at all, and it makes life tough for Aurodium players. But for most of the population the squish makes it a good chunk easier to climb by constantly pushing them up.
  • Dawnsinger wrote: »
    Drathuk916 wrote: »
    Dawnsinger wrote: »
    Drathuk916 wrote: »
    Drathuk916 wrote: »
    The squish is the answer. Instead of audorium being players ranked in the 60-80 percentile, as cg has said their goal is, it contains players in the 68-84 percentile.

    So that’s my original quote. To expand on it, cg stated in the beginning that their goal was to have 5 leagues evenly balanced in population. The individual division were targeted to be split at 10/25/30/25/10. The overall population has been relatively stable at 300-310k for a while which means if cg was meeting its stated goal each league should look like 6000/15000/18000/15000/6000.

    Now no league comes close to looking like that and it mainly has to do with the squish artificially forcing players towards division three. It’s true kyber has somewhat stabilized and you do have to play your way out of kyber vs being squished out. However, without spending too much time studying the math, it feels like the squish has made movement between leagues far more difficult than it should be. My gut instinct seems to be supported by the shrinking to almost disappearing entirely of the populations in division 1 and 5 in any league not just kyber.

    My apologies. I assumed you were the guy I replied to, I didn’t realize you were using my post to talk about a totally new (but admittedly interesting) topic.

    There definitely don’t appear to be using the squish to hone in on their initial targets for the size of each division. It is being used almost exclusively to shift players up from Carbonite and into the other leagues until you get to Aurodium. Without it the size of Bronzium would be absolutely mammoth. If anything they’re not being anyway near aggressive enough with the squish as evidence by the fact that Carbonite has grown so big it’s shifted all the leagues away from their stated targets. There just are too many people entering Carbonite every month and they aren’t shifting them out fast enough to keep up.

    If you do the math you’ll see it’s actually made it much easier to move up the leagues with the exception of Aurodium the past 6 months, but if you’re in Aurodium I can see why your gut would say it’s harder.

    I’m in k2. My personal experience is I don’t care cause I’m getting more crystals than I need to progress at the rate I desire. That said I feel the squish is too punishing in k2. You need to average around 5.2 wins per season to maintain your rating. I’d lower that to somewhere between 4.7-5 wins. I’m 19-11 over my past 30 matches and gained around 130-150 rating points that’s too low.

    Finally, I’m not entirely clear why you feel it’s easy to climb so I must be missing something. It looks like the three middle leagues have shrunk by at least 12 percent while carbonite has grown by over 33%.

    If you want to see the details you should check out the excellent work Taliana has done to show the details. Just search her name on the forums. But the short version is that Carbonite gets a massive boost every seasons from the squish (100s of extra SR) Bronzium gets a large boost and Chromium gets a small one. The reason they have to do this, and why Carbonite is still growing compare to the rest is that people are constant leaving and new players are joining. But they leave from all over and only join in Carbonite, so the lowest league is naturally getting bigger while everywhere else is shrinking.

    The squish definitely makes it tough to stay in high K2 or K1 at all, and it makes life tough for Aurodium players. But for most of the population the squish makes it a good chunk easier to climb by constantly pushing them up.

    I understand now. It’s actually a bleaker picture than I was thinking. I had assumed that each individual league had become basically its own ecosystem and only changing by having people choosing to leave it. That’s not great but assuming you’re correct, and I have no reason to doubt it, and movement is easier to at least audorium than even a greater amount of players are deciding to disengage from gac than I suspected. Now that’s somewhat offset by having a larger new player pool in this scenario but it’s not a good look.
  • Dawnsinger wrote: »
    Drathuk916 wrote: »
    Dawnsinger wrote: »
    Drathuk916 wrote: »
    Drathuk916 wrote: »
    The squish is the answer. Instead of audorium being players ranked in the 60-80 percentile, as cg has said their goal is, it contains players in the 68-84 percentile.

    So that’s my original quote. To expand on it, cg stated in the beginning that their goal was to have 5 leagues evenly balanced in population. The individual division were targeted to be split at 10/25/30/25/10. The overall population has been relatively stable at 300-310k for a while which means if cg was meeting its stated goal each league should look like 6000/15000/18000/15000/6000.

    Now no league comes close to looking like that and it mainly has to do with the squish artificially forcing players towards division three. It’s true kyber has somewhat stabilized and you do have to play your way out of kyber vs being squished out. However, without spending too much time studying the math, it feels like the squish has made movement between leagues far more difficult than it should be. My gut instinct seems to be supported by the shrinking to almost disappearing entirely of the populations in division 1 and 5 in any league not just kyber.

    My apologies. I assumed you were the guy I replied to, I didn’t realize you were using my post to talk about a totally new (but admittedly interesting) topic.

    There definitely don’t appear to be using the squish to hone in on their initial targets for the size of each division. It is being used almost exclusively to shift players up from Carbonite and into the other leagues until you get to Aurodium. Without it the size of Bronzium would be absolutely mammoth. If anything they’re not being anyway near aggressive enough with the squish as evidence by the fact that Carbonite has grown so big it’s shifted all the leagues away from their stated targets. There just are too many people entering Carbonite every month and they aren’t shifting them out fast enough to keep up.

    If you do the math you’ll see it’s actually made it much easier to move up the leagues with the exception of Aurodium the past 6 months, but if you’re in Aurodium I can see why your gut would say it’s harder.

    I’m in k2. My personal experience is I don’t care cause I’m getting more crystals than I need to progress at the rate I desire. That said I feel the squish is too punishing in k2. You need to average around 5.2 wins per season to maintain your rating. I’d lower that to somewhere between 4.7-5 wins. I’m 19-11 over my past 30 matches and gained around 130-150 rating points that’s too low.

    Finally, I’m not entirely clear why you feel it’s easy to climb so I must be missing something. It looks like the three middle leagues have shrunk by at least 12 percent while carbonite has grown by over 33%.


    The squish definitely makes it tough to stay in high K2 or K1 at all, and it makes life tough for Aurodium players. But for most of the population the squish makes it a good chunk easier to climb by constantly pushing them up.

    Yeah definitely. I used to be a K1 regular a while ago, but due to the constant squish it got impossible for me to stay there, let alone climb. So I've been a K2 regular for quite some time now. I did reach K1 again a few times, but only for like 1 week before dropping back to K2. As @Drathuk916 pointed out above, it's really tough for people in my rank, since we have to win more than we lose just to STAY where we are. So in order to climb we need to do significantly better. My peak rating used to be ~3750, but nowadays this would be impossible for me to reach. I fluctuate between 3400 and 3500 these days. And it's not that I'm suddenly performing worse than before, but the squish keeps dragging me down and keeps changing the "landscape" of the ranks. Back when I was at my peak 3750, I was like rank 5000 in the world (not 100% sure, but it shouldn't be far off. I will check it when I have some time because I believe I still have the screenshots). Someone at 3750 rating TODAY is probably like what, top 800? If anyone is at around 3750, please let me know what rank you are, I'm curious. So in short, the same SR is basically worth "more" nowadays in these divisions than it was back then, because of the squish.
  • TVF
    36519 posts Member
    Dawnsinger wrote: »
    Drathuk916 wrote: »
    Dawnsinger wrote: »
    Drathuk916 wrote: »
    Drathuk916 wrote: »
    The squish is the answer. Instead of audorium being players ranked in the 60-80 percentile, as cg has said their goal is, it contains players in the 68-84 percentile.

    So that’s my original quote. To expand on it, cg stated in the beginning that their goal was to have 5 leagues evenly balanced in population. The individual division were targeted to be split at 10/25/30/25/10. The overall population has been relatively stable at 300-310k for a while which means if cg was meeting its stated goal each league should look like 6000/15000/18000/15000/6000.

    Now no league comes close to looking like that and it mainly has to do with the squish artificially forcing players towards division three. It’s true kyber has somewhat stabilized and you do have to play your way out of kyber vs being squished out. However, without spending too much time studying the math, it feels like the squish has made movement between leagues far more difficult than it should be. My gut instinct seems to be supported by the shrinking to almost disappearing entirely of the populations in division 1 and 5 in any league not just kyber.

    My apologies. I assumed you were the guy I replied to, I didn’t realize you were using my post to talk about a totally new (but admittedly interesting) topic.

    There definitely don’t appear to be using the squish to hone in on their initial targets for the size of each division. It is being used almost exclusively to shift players up from Carbonite and into the other leagues until you get to Aurodium. Without it the size of Bronzium would be absolutely mammoth. If anything they’re not being anyway near aggressive enough with the squish as evidence by the fact that Carbonite has grown so big it’s shifted all the leagues away from their stated targets. There just are too many people entering Carbonite every month and they aren’t shifting them out fast enough to keep up.

    If you do the math you’ll see it’s actually made it much easier to move up the leagues with the exception of Aurodium the past 6 months, but if you’re in Aurodium I can see why your gut would say it’s harder.

    I’m in k2. My personal experience is I don’t care cause I’m getting more crystals than I need to progress at the rate I desire. That said I feel the squish is too punishing in k2. You need to average around 5.2 wins per season to maintain your rating. I’d lower that to somewhere between 4.7-5 wins. I’m 19-11 over my past 30 matches and gained around 130-150 rating points that’s too low.

    Finally, I’m not entirely clear why you feel it’s easy to climb so I must be missing something. It looks like the three middle leagues have shrunk by at least 12 percent while carbonite has grown by over 33%.


    The squish definitely makes it tough to stay in high K2 or K1 at all, and it makes life tough for Aurodium players. But for most of the population the squish makes it a good chunk easier to climb by constantly pushing them up.

    Yeah definitely. I used to be a K1 regular a while ago, but due to the constant squish it got impossible for me to stay there, let alone climb. So I've been a K2 regular for quite some time now. I did reach K1 again a few times, but only for like 1 week before dropping back to K2. As @Drathuk916 pointed out above, it's really tough for people in my rank, since we have to win more than we lose just to STAY where we are. So in order to climb we need to do significantly better. My peak rating used to be ~3750, but nowadays this would be impossible for me to reach. I fluctuate between 3400 and 3500 these days. And it's not that I'm suddenly performing worse than before, but the squish keeps dragging me down and keeps changing the "landscape" of the ranks. Back when I was at my peak 3750, I was like rank 5000 in the world (not 100% sure, but it shouldn't be far off. I will check it when I have some time because I believe I still have the screenshots). Someone at 3750 rating TODAY is probably like what, top 800? If anyone is at around 3750, please let me know what rank you are, I'm curious. So in short, the same SR is basically worth "more" nowadays in these divisions than it was back then, because of the squish.

    I got back to 3719 before the most recent squish and I was like #995.
    I need a new message here. https://discord.gg/AmStGTH
  • KDC99X
    750 posts Member
    Yeah, I've felt "stuck" in high Aurodium for awhile now - and now it's starting to make sense why I feel a tremendous sense of inertia when it comes to moving up!!

    Seriously, I'm really enjoying this conversation all around, it's very interesting and enlightening understanding the systemic dynamics.

    I also appreciate all the perspectives shared, including the ones contradicting my statement that "no one is really enjoying" GAC as-is. I still content that those voices appear to be in the minority, and I see a lot more discontent than contentment, but I'm glad to hear everyone sharing their experience with this new system, and I hope that all this conversation and feedback will legitimately be considered by CG towards improving the system further, for ALL players.
    Dawnsinger wrote: »
    Drathuk916 wrote: »
    Drathuk916 wrote: »
    The squish is the answer. Instead of audorium being players ranked in the 60-80 percentile, as cg has said their goal is, it contains players in the 68-84 percentile.

    So that’s my original quote. To expand on it, cg stated in the beginning that their goal was to have 5 leagues evenly balanced in population. The individual division were targeted to be split at 10/25/30/25/10. The overall population has been relatively stable at 300-310k for a while which means if cg was meeting its stated goal each league should look like 6000/15000/18000/15000/6000.

    Now no league comes close to looking like that and it mainly has to do with the squish artificially forcing players towards division three. It’s true kyber has somewhat stabilized and you do have to play your way out of kyber vs being squished out. However, without spending too much time studying the math, it feels like the squish has made movement between leagues far more difficult than it should be. My gut instinct seems to be supported by the shrinking to almost disappearing entirely of the populations in division 1 and 5 in any league not just kyber.

    My apologies. I assumed you were the guy I replied to, I didn’t realize you were using my post to talk about a totally new (but admittedly interesting) topic.

    There definitely don’t appear to be using the squish to hone in on their initial targets for the size of each division. It is being used almost exclusively to shift players up from Carbonite and into the other leagues until you get to Aurodium. Without it the size of Bronzium would be absolutely mammoth. If anything they’re not being anyway near aggressive enough with the squish as evidence by the fact that Carbonite has grown so big it’s shifted all the leagues away from their stated targets. There just are too many people entering Carbonite every month and they aren’t shifting them out fast enough to keep up.

    If you do the math you’ll see it’s actually made it much easier to move up the leagues with the exception of Aurodium the past 6 months, but if you’re in Aurodium I can see why your gut would say it’s harder.

  • TVF
    36519 posts Member
    Any "I see this more than that" regarding positive and negative opinions is inherently anecdotal.

    Negative opinions tend to be much louder.
    I need a new message here. https://discord.gg/AmStGTH
  • KDC99X
    750 posts Member
    edited March 2023
    Ok - thanks for that! So here's my comments, and I'm curious if they would alleviate your concerns:

    1) I never said the weighting system for roster composition is 1-for-1 (i.e. "I unlocked a GL, now I'm facing people with another GL than those I was previously facing") - I said it would need to be a distributed range around your skill range/roster composition. So for instance, if I can reasonably compete with opponents with up to, say, 3 more GLs than me, then that range would include those players. And players with lower weighted rosters, but are capable of competing against me (say players with 2-3 GLs less than me, if they're good). Adding an additional GL to my roster wouldn't immediately just tack on another GL to all my opponents - we're talking about something like a Gaussian distribution around a mean. Having one more GL may or may not make a big difference in your capabilities and win/loss ratio, depending on the other supporting toons you have in your roster, and whether you use your GL on defense or offense, what mods you have, and so on. So again, it's a combination of this proposed "weighting system" AND skill rating, which prevents it from being either homogeneous in the roster composition of opponents you face, or it being a "zero-sum" game. You do well with what you have, you can still progress and move up the ladder. You're just guaranteed more interesting and competitive matches, tailored to your skill level and the tools you have at your disposal.

    2) Note that our current system is a "zero-sum" game anyway on the macro scale. For everyone that goes up in SR, others go down by an equivalent amount. If I understand what Dawnsinger, Taliana, and others have shared on how the squish works.

    3) There are a lot of ways to quantify these things, but probably the best way would be to just analyze the existing data sets. There's mounds of data on successful counters, etc. Every bracket, I pull up my bracket comparison on swgoh.gg and look at all my opponents, comparing GLs, relic/gear levels, zetas, omicrons, key legendary toons, capital ships, mods, etc. I've been doing this long enough, that this overview generally gives me a very good idea of what I'm up against, and the rough probability of success in various matchups. So just analyze the existing wins and losses of players, and successful/failed counters, etc., against that data, to come up with the initial tuning coefficients for how to weight the different variables. It won't be perfect because there's a lot of confounding factors (for example, players who drop down intentionally to face easier opponents), but it would be a great first stab, and the coefficients could be further tweaked over time. I imagine a machine learning algorithm would probably be great for this, but I'm not a ML expert.

    So, if this system could be implement in a way that eliminates your concerns, would you still be against it?

    Edit: typo
    Screerider wrote: »
    KDC99X wrote: »
    Why is that?

    Or more specifically, what would be the negative outcomes you would want to avoid in such a schema?
    Screerider wrote: »
    I don't want roster to be considered. I like it as-is.
    The most notable negative outcome is being penalized for improving your roster. Yay, I unlocked a GL, and now I'm facing people with another GL than those I was previously facing. It's a zero-sum game. Why bother?

    Also, I just don't think there's a way to quantify a roster in any meaningful way. What's a GL worth? A An Omicron? A Zeta? A Datacron? Each Relic Level? It's a fools errand. Better to have them be "quantified" by whether it helps you win or not, which is what the current system does.

  • KDC99X
    750 posts Member
    Sure, but many times negative opinions are louder because they are also in the majority. In every sphere of the community, from here, to in-game, to every content creator out there and the endless comments on their videos, I see most people overwhelmingly expressing dissatisfaction. Heck even in this thread, if we were to take a vote, there's a great deal more people saying they're unhappy with the current system than happy.

    That doesn't mean nothing, and to dismiss it out-of-hand just seem an odd stance to take.
    TVF wrote: »
    Any "I see this more than that" regarding positive and negative opinions is inherently anecdotal.

    Negative opinions tend to be much louder.
  • KDC99X wrote: »
    Sure, but many times negative opinions are louder because they are also in the majority. In every sphere of the community, from here, to in-game, to every content creator out there and the endless comments on their videos, I see most people overwhelmingly expressing dissatisfaction. Heck even in this thread, if we were to take a vote, there's a great deal more people saying they're unhappy with the current system than happy.

    That doesn't mean nothing, and to dismiss it out-of-hand just seem an odd stance to take.
    1) by this rationale, every restaurant, hotel I ever check reviews for must suck, because the majority of them are complaints.

    2) please consider putting your quote UNDER the post you are quoting instead of above. It’s very annoying having to delete quotes from your posts when replying to you so that we are quoting your words, not the person you quoted’s words.
  • Screerider
    1343 posts Member
    KDC99X wrote: »
    So again, it's a combination of this proposed "weighting system" AND skill rating

    So, if this system could be implement in a way that eliminates your concerns, would you still be against it?
    Yup. For reasons previously noted.

  • Screerider
    1343 posts Member
    Except the current system doesn’t do this. I regularly play against opponents where nothing I do will help me “win or not”. On average this happens for me with every 2 out of 3 GAC opponents.
    Improving your roster increases your odds of winning more matches. Plain and simple.

  • Screerider wrote: »
    Except the current system doesn’t do this. I regularly play against opponents where nothing I do will help me “win or not”. On average this happens for me with every 2 out of 3 GAC opponents.
    Improving your roster increases your odds of winning more matches. Plain and simple.

    I’m actually saying the exact opposite. Improving my roster doesn’t change the odds at all of winning most of my matches. That’s the problem.
  • Screerider
    1343 posts Member
    I’m actually saying the exact opposite. Improving my roster doesn’t change the odds at all of winning most of my matches. That’s the problem.
    Sounds like something to work for.
  • Screerider wrote: »
    Except the current system doesn’t do this. I regularly play against opponents where nothing I do will help me “win or not”. On average this happens for me with every 2 out of 3 GAC opponents.
    Improving your roster increases your odds of winning more matches. Plain and simple.

    Improving your roster at the same rate as everyone else ranked around you DOES NOT improve your chances of winning. You only tread water, with a slight variance based on what you choose to pursue in comparison to those around you.

    This is by design, so people are encouraged to spend to separate themselves from the pack. If you improve faster than those around you, then it’ll improve your chances at winning. That’s the “plain and simple” part you seem to be overlooking.

    But if you improve at the same rate as everyone else, you’re going to have to be happy with the status quo. Not that there’s anything wrong with that - it’s just the way the game is designed.
  • I definitely think that GAC could encourage/reward roster growth more if they found a way to adjust matchmaking so that the lower leagues didn’t have such massive mismatches. If a significant portion your matches are coming down to whether your opponent bothers or not and it’s a gap of 3 or 4 GLs and several million GP, that’s not really a huge incentive to grow your roster.

    I am curious how prevalent that actually is though. I see the complaint that “most of matches I have no chance to win” a lot. But in the handful of times I’ve seen someone’s actual matches, it’s been a couple matches a season where they were against unbeatable opponents (say 3+ GLs), and a bunch of matches where just having an extra GL would have given them a much better shot. If you’ve reached a point where you’re mostly matched against opponents with 2 more GLs than you, then growing your roster by one more GL definitely will give you a much better chance to win, which is what CG wants.

    I’d love to see some rosters where most of the matches are against truly unbeatable rosters though. The kind of rosters were adding another GL still wouldn’t have given you any chance. I’ve heard about it a bunch but never really seen it in action.
  • Screerider
    1343 posts Member
    TheDude420 wrote: »
    Screerider wrote: »
    Except the current system doesn’t do this. I regularly play against opponents where nothing I do will help me “win or not”. On average this happens for me with every 2 out of 3 GAC opponents.
    Improving your roster increases your odds of winning more matches. Plain and simple.

    Improving your roster at the same rate as everyone else ranked around you DOES NOT improve your chances of winning. You only tread water, with a slight variance based on what you choose to pursue in comparison to those around you.

    This is by design, so people are encouraged to spend to separate themselves from the pack. If you improve faster than those around you, then it’ll improve your chances at winning. That’s the “plain and simple” part you seem to be overlooking.

    But if you improve at the same rate as everyone else, you’re going to have to be happy with the status quo. Not that there’s anything wrong with that - it’s just the way the game is designed.
    No doubt.

  • Dawnsinger wrote: »
    I definitely think that GAC could encourage/reward roster growth more if they found a way to adjust matchmaking so that the lower leagues didn’t have such massive mismatches. If a significant portion your matches are coming down to whether your opponent bothers or not and it’s a gap of 3 or 4 GLs and several million GP, that’s not really a huge incentive to grow your roster.

    I am curious how prevalent that actually is though. I see the complaint that “most of matches I have no chance to win” a lot. But in the handful of times I’ve seen someone’s actual matches, it’s been a couple matches a season where they were against unbeatable opponents (say 3+ GLs), and a bunch of matches where just having an extra GL would have given them a much better shot. If you’ve reached a point where you’re mostly matched against opponents with 2 more GLs than you, then growing your roster by one more GL definitely will give you a much better chance to win, which is what CG wants.

    I’d love to see some rosters where most of the matches are against truly unbeatable rosters though. The kind of rosters were adding another GL still wouldn’t have given you any chance. I’ve heard about it a bunch but never really seen it in action.

    That's fair. Here is my swgoh GAC History https://swgoh.gg/p/297932863/gac-history/

    I currently have 3.1 GP and no GLs. Here are my last 9 opponents' GP and # of GLs which I know isn't a complete picture but gives you an idea.

    1) 7.5 GP, 4 GLs
    2) 6.7 GP, 2 GLs
    3) 7.9 GP, 2 GLs
    4) 6.8 GP, 1 GL
    5) 4.3 GP, 1 GL
    6) 7.5 GP, 1 GL
    7) 8.4 GP, 5 GLs
    8) 4.1 GP, 1 GL
    9) 7.9 GP, 1 GL

    Of these matches, the only ones where I felt like what I did mattered was #5 and #8. I agree that having 1 additional GL would have helped in some of these matches, but not all. It isn't just the GL difference, but having enough full relic squads to fill out defense and still have good offensive squads which generally comes down to GP.

    Still, I do tend to win about half of my matches which keeps me settled between chromium 1 and 2. Of the 9 matches above I won 4. #s 5 and 8 were genuinely fun matchups where my opponent showed up and I was able to get the win. But I also won #s 2 and 9 because my opponent didn't try. I lost battles 1, 3, 4, 6, and 7 because my opponent showed up. For me, the current system doesn't create a lot of fun matchups.
  • KDC99X
    750 posts Member
    But that contradicts your assertion that the current system accounts for roster improvement.
    Screerider wrote: »
    TheDude420 wrote: »
    Screerider wrote: »
    Except the current system doesn’t do this. I regularly play against opponents where nothing I do will help me “win or not”. On average this happens for me with every 2 out of 3 GAC opponents.
    Improving your roster increases your odds of winning more matches. Plain and simple.

    Improving your roster at the same rate as everyone else ranked around you DOES NOT improve your chances of winning. You only tread water, with a slight variance based on what you choose to pursue in comparison to those around you.

    This is by design, so people are encouraged to spend to separate themselves from the pack. If you improve faster than those around you, then it’ll improve your chances at winning. That’s the “plain and simple” part you seem to be overlooking.

    But if you improve at the same rate as everyone else, you’re going to have to be happy with the status quo. Not that there’s anything wrong with that - it’s just the way the game is designed.
    No doubt.

    In the proposed system, those reasons are negligible or eliminated outright, so it seems your stance is more emotional in nature, which is fine, but I see that you won’t be persuaded.

    Thanks for sharing your opinion, though!
    Screerider wrote: »
    KDC99X wrote: »
    So again, it's a combination of this proposed "weighting system" AND skill rating

    So, if this system could be implement in a way that eliminates your concerns, would you still be against it?
    Yup. For reasons previously noted.

  • Screerider
    1343 posts Member
    KDC99X wrote: »
    But that contradicts your assertion that the current system accounts for roster improvement.
    Not at all. I improve my roster, odds of winning increase. Opponent improves their roster, their odds of winning increase. So, yes, it's sometimes a wash. But not always. At least, you're less likely to fall.

    It encourages roster growth, so you (1) won't be left behind and (2) have a chance to climb. It's great really, compared to any system that attempts to tie rosters into matchmaking.

    I don't see how that's emotional, so trying to demean my opinion as not being thought-out is distasteful.
  • Gifafi
    6017 posts Member
    Bc ain’t nobody got time for that. The good news is vs me you’ll have about a 60/40 chance I’ll be too busy and/or forget about gac and you’ll get an easy win. If I do play then rip. What fun!
    Maybe End Game isn't for you
  • nfidel2k
    559 posts Member
    There’s no easy way to determine roster strength, even using things like GL count, omis, speed mods, etc. It would be far easier to deal with the problem by splitting the leagues and imposing GP minimums and maximums for league entry. Keep the skill ratings for determining division, but kill it for league advancement. No need for GP to be a part of matchmaking if it already limits your opponents by league entry.

    There are problems with this though, including that rewards would need to be restructured. I’m not exactly sure this would be better than the current system, but it is the easiest way to deal with large accounts dropping.
    Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for life.
  • KDC99X
    750 posts Member
    That’s hyperbole, and an inaccurate comparison. Most review sites have a wide range of reviews, and you can look at the overall average and distribution. If most review are skewed negative (generally, 3 stars or lower on a 5 star system), it indicates to look more closely at why, and when reading negative (and positive) reviews, you can usually pick out completely unreasonable complaints for legitimate concerns. Even if most are positive, it’s always good look more closely into the actual verbiage used, which reveals a lot more about whether those reviews are worth taking at face value, or with a huge grain of salt.
    KDC99X wrote: »
    Sure, but many times negative opinions are louder because they are also in the majority. In every sphere of the community, from here, to in-game, to every content creator out there and the endless comments on their videos, I see most people overwhelmingly expressing dissatisfaction. Heck even in this thread, if we were to take a vote, there's a great deal more people saying they're unhappy with the current system than happy.

    That doesn't mean nothing, and to dismiss it out-of-hand just seem an odd stance to take.
    1) by this rationale, every restaurant, hotel I ever check reviews for must suck, because the majority of them are complaints.

    2) please consider putting your quote UNDER the post you are quoting instead of above. It’s very annoying having to delete quotes from your posts when replying to you so that we are quoting your words, not the person you quoted’s words.

  • KDC99X wrote: »
    That’s hyperbole, and an inaccurate comparison. Most review sites have a wide range of reviews, and you can look at the overall average and distribution. If most review are skewed negative (generally, 3 stars or lower on a 5 star system), it indicates to look more closely at why, and when reading negative (and positive) reviews, you can usually pick out completely unreasonable complaints for legitimate concerns. Even if most are positive, it’s always good look more closely into the actual verbiage used, which reveals a lot more about whether those reviews are worth taking at face value, or with a huge grain of salt.
    Any chance you’ll move on point 2?

    It’s so frustrating having to delete quotes every time, I’m genuinely considering putting you on ignore. You are the only forumer who uses quotes this way.
  • Awesome, (well those matchups aren’t awesome, but I appreciate the data). So I’m curious, for you personally if there was some form of banding so your matchups were made of opponents that only had 2 or less GLs, would it be more tempting to spend to expand your roster? Or would the high GP/low GL accounts be demoralizing as well? Just curious which parts of those matchups feel most unfair for folks like you that are punching above their weight class.

    Dawnsinger wrote: »
    I definitely think that GAC could encourage/reward roster growth more if they found a way to adjust matchmaking so that the lower leagues didn’t have such massive mismatches. If a significant portion your matches are coming down to whether your opponent bothers or not and it’s a gap of 3 or 4 GLs and several million GP, that’s not really a huge incentive to grow your roster.

    I am curious how prevalent that actually is though. I see the complaint that “most of matches I have no chance to win” a lot. But in the handful of times I’ve seen someone’s actual matches, it’s been a couple matches a season where they were against unbeatable opponents (say 3+ GLs), and a bunch of matches where just having an extra GL would have given them a much better shot. If you’ve reached a point where you’re mostly matched against opponents with 2 more GLs than you, then growing your roster by one more GL definitely will give you a much better chance to win, which is what CG wants.

    I’d love to see some rosters where most of the matches are against truly unbeatable rosters though. The kind of rosters were adding another GL still wouldn’t have given you any chance. I’ve heard about it a bunch but never really seen it in action.

    That's fair. Here is my swgoh GAC History https://swgoh.gg/p/297932863/gac-history/

    I currently have 3.1 GP and no GLs. Here are my last 9 opponents' GP and # of GLs which I know isn't a complete picture but gives you an idea.

    1) 7.5 GP, 4 GLs
    2) 6.7 GP, 2 GLs
    3) 7.9 GP, 2 GLs
    4) 6.8 GP, 1 GL
    5) 4.3 GP, 1 GL
    6) 7.5 GP, 1 GL
    7) 8.4 GP, 5 GLs
    8) 4.1 GP, 1 GL
    9) 7.9 GP, 1 GL

    Of these matches, the only ones where I felt like what I did mattered was #5 and #8. I agree that having 1 additional GL would have helped in some of these matches, but not all. It isn't just the GL difference, but having enough full relic squads to fill out defense and still have good offensive squads which generally comes down to GP.

    Still, I do tend to win about half of my matches which keeps me settled between chromium 1 and 2. Of the 9 matches above I won 4. #s 5 and 8 were genuinely fun matchups where my opponent showed up and I was able to get the win. But I also won #s 2 and 9 because my opponent didn't try. I lost battles 1, 3, 4, 6, and 7 because my opponent showed up. For me, the current system doesn't create a lot of fun matchups.

  • KDC99X
    750 posts Member
    Easiest yes, BUT that doesn’t eliminate the incentive to keep your GP low to stay in a favorable bracket, so I don’t see them going back that route.

    A ML model would probably do the job nicely, at least providing a good initial set of tuning coefficients for the weighting algorithm.
    nfidel2k wrote: »
    There’s no easy way to determine roster strength, even using things like GL count, omis, speed mods, etc. It would be far easier to deal with the problem by splitting the leagues and imposing GP minimums and maximums for league entry. Keep the skill ratings for determining division, but kill it for league advancement. No need for GP to be a part of matchmaking if it already limits your opponents by league entry.

    There are problems with this though, including that rewards would need to be restructured. I’m not exactly sure this would be better than the current system, but it is the easiest way to deal with large accounts dropping.

  • KDC99X
    750 posts Member
    You don’t need to delete anything - older quotes past the latest are automatically collapsed.
    KDC99X wrote: »
    That’s hyperbole, and an inaccurate comparison. Most review sites have a wide range of reviews, and you can look at the overall average and distribution. If most review are skewed negative (generally, 3 stars or lower on a 5 star system), it indicates to look more closely at why, and when reading negative (and positive) reviews, you can usually pick out completely unreasonable complaints for legitimate concerns. Even if most are positive, it’s always good look more closely into the actual verbiage used, which reveals a lot more about whether those reviews are worth taking at face value, or with a huge grain of salt.
    Any chance you’ll move on point 2?

    It’s so frustrating having to delete quotes every time, I’m genuinely considering putting you on ignore. You are the only forumer who uses quotes this way.

  • KDC99X
    750 posts Member
    If you refuse to change your opinion in the face of contradictory evidence, then it’s an emotionally-held position.

    That’s not demeaning, it’s just acknowledging that no matter how much I say that the concerns you raised won’t be a problem, you’re not going to change your mind. And like I said before, that’s fine.
    Screerider wrote: »
    KDC99X wrote: »
    But that contradicts your assertion that the current system accounts for roster improvement.
    Not at all. I improve my roster, odds of winning increase. Opponent improves their roster, their odds of winning increase. So, yes, it's sometimes a wash. But not always. At least, you're less likely to fall.

    It encourages roster growth, so you (1) won't be left behind and (2) have a chance to climb. It's great really, compared to any system that attempts to tie rosters into matchmaking.

    I don't see how that's emotional, so trying to demean my opinion as not being thought-out is distasteful.

  • nfidel2k
    559 posts Member
    KDC99X wrote: »
    Easiest yes, BUT that doesn’t eliminate the incentive to keep your GP low to stay in a favorable bracket, so I don’t see them going back that route.

    Yes, it’s a step backwards slightly in that regards. But the leagues would be an advance-only system tied to account growth, so it would encourage it as well.

    One of the problems I see is that neither it nor the current system deals with account separation and account saturation/compression. Separation meaning the GP difference between a new account and an old account, and saturation being that as time progresses, more and more accounts will reach a viable Kyber level, which will expand down the leagues raising the bar. This leads to new accounts eventually being relegated to a maximum of Chromium because they can’t outgrow the accounts above them. Personally I think the ladder system needs to grow with the game, so as accounts expand, more leagues are generated. It can’t keep being compressed into a fixed number of leagues and divisions.
    Give a man a fish and you feed him for a day. Teach a man to fish and you feed him for life.
Sign In or Register to comment.